Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/albiol2004/trio-agent-loopWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/albiol2004/trio-agent-loop/trio-productionize)<a href="https://agentmods.dev/commands/albiol2004/trio-agent-loop/trio-productionize"><img src="https://agentmods.dev/badge/commands/albiol2004/trio-agent-loop/trio-productionize.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00014 | $0.00252 |
| Opus 5 | $0.00007 | $0.00126 |
| Sonnet 5 | $0.00003 | $0.00050 |
| Haiku 4.5 | $0.00001 | $0.00025 |
Grade A, and why
trio-productionize scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Omp Trio productionize
Read and follow $PZ_HOME/command.md; it is the canonical productionize procedure and owns the graph, driver, batching, recording, and close-out rules.
Dispatch table
executor: scoutprobe nodes: usetask(agent:"scout")with the batch briefing and known context.executor: assessor:standardandexecutor: assessor:highjudgment nodes: use the defaulttaskagent with the batch briefing and recorded probe evidence.executor: usernodes: ask through theasktool, then record the user's decision verbatim as evidence.
Do not dispatch from raw plan.json; give agents their batch briefing and enforce the command's context caps. The orchestrator records each result with the driver's record-batch.
Delivery
Every agent MUST write its verdict array to pz-run/results/<batch-stem>.json before composing its final reply. Before its formal yield, it MUST also hub-DM the raw verdict JSON array to Main; record from the file or DM, accepting partial arrays and re-dispatching only gaps.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago Changed · -20 lines · -34 tokens per session c36702a70fdc
- 7d ago First seen · 40 lines · 48 tokens per session scan A fea21dcac77b
trio-productionize is a command published in the GitHub repository albiol2004/trio-agent-loop (2 stars, last pushed 7d ago), licensed MIT. It adds 14 tokens to every session and 252 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
fire-0-orient
Orient on an existing project - understand what's here and what's next.
task-verify
PitWay: Run an inprogress task's approved verification command and persist evidence.
merged
Post-merge cleanup: confirm the PR actually merged, move the Jira ticket to the project's post-review state, sync the default branch, and delete the merged feature branch.
lfe-plan-critique
Run a 5-lens pre-build critique of the approved active plan before the Builder starts. Acts as the Architect persona, read-only on src/. Writes .plans/plancritique.md. Use immediately after Brain approves activeplan.md.
install
This page explains how the livewiki install CLI command detects installed coding agents and safely configures the livewiki MCP server entry, hook templates, shared skills, and optional pointer files.
plan-feature
Plan and implement a feature through the VibeFrame model (context → declare → implement → link → validate).